Reverberant audio source separation using partially pre-trained nonnegative matrix factorization

Fakhry, Mahmoud; Svaizer, Piergiorgio; Omologo, Maurizio

doi:10.1109/iwaenc.2014.6954301

Cited by 4 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, the reverberation frequency basis matrix W R was initialized (pre-trained) before NMF update to making it stable and robust with environmental fluctuations. This concept is widely employed in speech, audio, and acoustic signal processing [20,21,22], but we apply it to the sonar reverberation suppression method for the first time. Second, a power normalization scheme was adopted to compensate for the imbalance between the target echo and reverberation signal.…”

Section: The Proposed Nmf-based Reverberation Suppression Methods Wit...mentioning

confidence: 99%

See 1 more Smart Citation

Reverberation Suppression Method for Active Sonar Systems Using Non-Negative Matrix Factorization With Pre-Trained Frequency Basis Matrix

Kim

Lee

2021

IEEE Access

View full text Add to dashboard Cite

In active sonar systems, detection always suffers from reverberation interference from multiple scatterers in oceanic environments; therefore, numerous studies have been conducted on reverberation suppression. Recently, a non-negative matrix factorization (NMF)-based method was proposed and successfully applied to reverberation suppression. However, the conventional NMF-based method makes convergence challenging because the frequency basis matrix is initialized without considering reverberation characteristic information from oceanic environments. To solve these problems, We propose an improved NMF-based reverberation suppression method adopting a pre-trained reverberation basis matrix and modified sparse update rule. The proposed method is evaluated by analyzing simulation and sea experiment data and the study confirmed that the detection performance was improved compared to the conventional method under various signal-to-reverberation ratio conditions. Several topics are also discussed to analyze the proposed method in detail.

show abstract

Section: The Proposed Nmf-based Reverberation Suppression Methods Wit...mentioning

confidence: 99%

“…This scheme can be implemented by learning the reverberation frequency basis matrix W R in advance. Notably, this scheme of pre-trained basis matrix is widely used in speech, audio, and acoustic signal processing [20,21,22].…”

Section: B Pre-training the Frequency Basis Matrix With Power Normali...mentioning

confidence: 99%

Reverberation Suppression Method for Active Sonar Systems Using Non-Negative Matrix Factorization With Pre-Trained Frequency Basis Matrix

Kim

Lee

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

Estimation of the spatial information in Gaussian model based audio source separation using weighted spectral bases

Fakhry

Svaizer

Omologo

2016

2016 24th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

Audio source separation using a redundant library of source spectral bases for non-negative tensor factorization

Fakhry

Svaizer

Omologo

2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Reverberant audio source separation using partially pre-trained nonnegative matrix factorization

Cited by 4 publications

References 19 publications

Reverberation Suppression Method for Active Sonar Systems Using Non-Negative Matrix Factorization With Pre-Trained Frequency Basis Matrix

Reverberation Suppression Method for Active Sonar Systems Using Non-Negative Matrix Factorization With Pre-Trained Frequency Basis Matrix

Estimation of the spatial information in Gaussian model based audio source separation using weighted spectral bases

Audio source separation using a redundant library of source spectral bases for non-negative tensor factorization

Contact Info

Product

Resources

About